关键词: 概率分布, 概率分布, 重现期, Gumbel函数

Abstract: In order to improve the service of forecasting wind in Zhejiang coastal, this paper evaluated the forecast performance of the ECMWF fine mesh 10 m wind from January 1, 2012 to December 31, 2014, and interpolated them into the coastal stations comparing with the observed data. The results show that under the influence of cold air, the mean error of Zhejiang coastal is negative, and the distribution and variation trend of the absolute error are consistent with the mean error; under the influence of typhoon, the mean error of 24 h forecast time is positive, while it gradually turns to negative with the forecast time increasing, the absolute error of Yushan fishery, Wentai fishery and Zhouwai fishery is larger than other fisheries. In addition, the correlation coefficient of the forecast value and the observed data decreases with forecast time increasing, and the mean value of the correlation coefficient decreases with the increase of altitude. the forecast value of ECMWF fine mesh model is bigger than the observed data for Daishan and Longshancun, and smaller for Langang.; the dispersion of the mean error increases with forecast time increasing.